THIRD SECTION

CASE OF M.O. v. SWITZERLAND

( Application no. 41282/16 )

STRASBOURG

20 June 2017

FINAL

20/09/2017

This judgment has become final under Article 44 § 2 of the Convention. It may be subject to editorial revision.

In the case of M.O. v. Switzerland ,

The European Court of Human Rights ( Third Section ), sitting as a Chamber composed of:

Helena Jäderblom, President, Branko Lubarda, Helen Keller, Pere Pastor Vilanova, Alena Poláčková, Georgios A. Serghides, Jolien Schukking, judges, and Stephen Phillips , Section Registrar ,

Having deliberated in private on 30 May 2017,

Delivers the following judgment, which was adopted on that date:

PROCEDURE

1 . The case originated in an application (no. 41282/16) against the Swiss Confederation lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by an Eritrean national, Mr M.O. (“the applicant”), on 13 July 2016 .

2 . The applicant, who had been granted l egal aid, was represented by Mr R. Schuler and Ms A. Stettler, lawyer s practising in Zürich. The Swiss Government (“the Government”) were represented by their Deputy Agent, Mr A. Scheidegger , of the Federal Office of Justice .

3 . The applicant alleged that his expulsion to Eritrea would give rise to a violation of Articles 3 and 4 of the Convention .

4 . On 22 July 2016 the Vice-President of the Section to which the case had been allocated decided to apply Rule 39 of the Rules of Court, indicating to the Government that the applicant should not be expelled to Eritrea for the duration of the proceedings before the Court, and grant ed priority to the application under Rule 41 of the Rules of Court, and anonymity to the applicant under Rule 47 § 4 of the Rules of Court .

5 . On the same day, the application was communicated to the Government .

THE FACTS

I. THE CIRCUMSTANCES OF THE CASE

6 . The applicant is an Eritrean national and was born in 1990 . He grew up in Eritrea and currently lives in Switzerland .

7 . The applicant entered Switzerland illegally on 23 June 2014 and applied for asylum the next day . He was heard in person three times by the Swiss authorities responsible for asylum and migration (until 31 December 2014 the Bundesamt für Migration , renamed with effect from 1 January 2015 as the Staatssekretariat für Migration , SEM – hereafter “ the State Secretariat for Migration”). During all three hearings an interpreter was present and the record was translated for the applicant prior to his signing it.

8 . The first hearing was a summary interview at the Centre for Reception and Procedure ( Empfangs- und Verfahrenszentrum ) in Kreuzlingen on 1 July 2014. T he applicant stated that he had not been allowed to continue at school beyond the eight h grade because he had failed the national admission exams for secondary school , and had been summoned for military service, which he had tried to avoid. Once he had reported for duty, he had tried to escape but had been caught. He stated that he had been beaten and s ubsequently imprisoned in Wi ’ a in conditions of very poor hygien e . He was unable to re call the exact dates of his imprisonment, but stated that he had been imprisoned from June 2012 to September 2013. He had managed to escape from prison one night when the guards were asleep. After staying in Eritrea for one more week, he had left the country on foot on 3 October 2013 , crossing the border at Mereb. He had been picked up by Ethiopian soldiers the next day.

9 . In order to support his account, the applicant submitted copies of his student card and a card showing that he was a church deacon , as well as the original of a card used for food distribution in Italy. He stated that he had been issued with an identity card in Eritrea in 2010, but had had to hand it in in Ethiopia. The applicant further stated that he was married and had a son born in October 2012.

10 . The second, more detailed , hearing took place at the office of the State Secretariat for Migration in Berne on 11 March 2015. A member of a non-governmental organisation was present as a neutral witness , in order to guarantee the fairness of the hearing. He had the opportunity to add comments at the end of the record of the hearing in the event that he had witnessed any irregularities, but did not note down any such observations.

11 . The applicant again gave an account of the alleged events in Eritrea leading up to his escape. This time h e stated that he had been imprisoned in Wi ’ a from March to October 2013 following his attempt to escape from the military. When confronted with the discrepancy in compar ison with h is previous account in that regard, he stated that he might have made a mistake in the first interview. When asked about the conditions of his detention, the applicant claimed that it had been very dirty and very hot , and that he had been locked up all the time. Other than beatings suffered due to his attempt to escape, there had been no particular incidents. He did not recall any rules other than fixed meal times and being brought outside in the mornings and evenings to relieve himself.

12 . When asked about his military training, t he applicant claimed that he had only been there for a very brief period of time prior to his attempted escape . He had n ot learned how to use weapons . He could neither provide a name of his superior n or his unit nor his military number.

13 . When q uestioned about his departure from Eritrea, the applicant described leaving A . (a village) on foot at around 6 p.m., together with a person living in the neighbourhood who knew the area . They had walked towards Mereb , but had lost the ir way as it was night -time . They had been very afraid a nd also hungry and thirsty. The re w ere lots of thorn bushes in their way and they had heard the howling of hyenas. They had not been sure whether they had reached Ethiopian territory until soldiers who spoke Amharic had apprehended them at around 3 a.m . When informed by the interviewer that his account in relation to his departure m ight not be considered credible, which would result in the conclusion that he had left the country legally, t he applicant said that he could not gi ve more details about the departure because he did not know the area well and had been following the person with whom he had fled.

14 . T h e applicant claimed t hat he had alwa ys been in good health.

15 . In order to support his account, the applicant submitted the originals of his marriage certif icate of 2010 and his son ’ s baptism certificate of 2012 , as well as an attestation that he was a church deacon. He claimed that these documents had been sent to him from Eritrea.

16 . The applicant was heard for a third time by the State Secretariat for Migration on 29 January 2016. Again, a member of a non-governmental organisation was present as a neutral witness , who did not note down any observations relating to irregularities .

17 . The interviewer explicitly advised the applicant that the interview was t aking place to give him another opportunity to describe his departure from Eritrea , and that the account he had so far provided in this regard would probably be dismissed as not credible, which would result in the conclusion that he had left the country legally.

18 . The applicant responded that he had spent two days a t his parents ’ home in A. after escaping from prison. His family had contacted a people smuggler from the Mereb area. He and the smuggler had left A . on foot at around 6 p.m. When confronted with his previous account that he had left with a person from the neighbourhood, the applicant clarified that the person he had travelled with (the smuggler) came from a neighbouring village. During the night the smuggler had told him that he had taken many pe ople across the border , but that he and the applicant had already been walking for too long a time, which meant that they must have lost the ir way. Th ey had become dis orientat ed and had only realised that they had crossed the border when they had been apprehended by Ethiopian soldiers at around 4 a.m. who had spoken to them in Amharic .

19 . On 8 March 2016 the S tate Secretariat for Migration rejected the applicant ’ s asylum request and ordered his departure from Switzerland. It found that his account was not credible , and concluded that, having failed to prove or credibly demonstrate his refugee status pursuant to section 7 of the Asylum Act, the applicant was not a refugee as defin ed in section 3 of the Asylum Act.

20 . The State Secretariat for Migration pointed out that the applicant had stated in the first hearing that he had been detained for a period of one year and three months , from June 2012 to September 2013, whereas in the second hearing he had said that he had been detained for seven months, from March to October 2013. It added that the applicant had also contradicted himself a number of times in the second hearing in relation to the commencement of his military training and the end of his schooling. Furthermore, his account as regards his time in prison and his escape from prison lacked substance , and he could not provide details as to what he had learned during his military training.

21 . The State Secretariat for Migration also found that the applicant ’ s submissions regarding his illegal exit from Eritrea were not credible. Despite being asked several times to provide a detailed account of his departure or specific events in that connection, the applicant ’ s statement in that regard was superficial and limited to phrases. The State Secretariat for Migration argued that, particular ly in relation to the hours during which the applicant and the smuggler had lost the ir way, it could legitimately be expected that the applicant would provide a substantiated account , which he had failed to do . Nor had he provide d a consistent account in that regard . Moreover , he had made contradictory statements as to how long he had stayed at home between escaping from prison and leaving the country .

22 . On 14 April 2016 the applicant lodged an appeal against that decision before the Federal Administr ative Court. He submitted that the authorities had initially refrained from drafting him because of his role as a deacon – the fact that he was a deacon not having been contested by the asylum authority . He referred to a report of the European Asylum Support Office (EASO) , which stated that the relevant authorities had, at times, deferred the draft of clerics, until a change in practice in 2010 had led to a stricter approach (see paragraph 49 below ) . The applicant submitted that he had had to undergo his military training in Wi ’ a, a place to which, according to a report by the US Department of State (see paragraph 5 1 ) , students with poor grades were typically assigned. Shortly after reporting for duty, he had attempted to escape. He had been caught and, as a result, detained in March 2013. He had managed to escape from detention in September 2013 after seven months. Fearing that he would be detained once again or forced to perform military service , he had decided in early October 2013 to leave the country illegally. His family had organised a smuggler. They had fled during the night and, while attempting to cross the border at Mereb, had lost the ir way. They had been apprehended by Ethiopian soldiers early in the morning of 4 October 2013 and taken to Endabaguna. One week later, they had been transferred to the Hitsas refugee camp.

23 . With regard to the alleged discrepancies relating to his schooling, the applicant submitted that he had difficulties with dates concerning the duration of his schooling and his age when he had started and left school, but emphasised that he had consistently state d that he had left school after eighth grade , having failed the exams to move on to secondary school, and referred to a report by EASO regarding the national examination at the end of eighth grade (see paragraph 49 below ). Also, during the hearing he had corrected himself in relation to the commencement of his military training. F urthermore, it was comprehensible that he could not give a detailed account about things which he had learned during that training, given that he had attempted to escape almost imm ediate ly after reporting for duty.

24 . The applicant claimed that the discrepancies between his statements in the first and the second hearing s concerning the duration of his detention were due to his poor level of education. H e emphasised that his statements as regards the time of the end of his detention, September 2013, were consistent. In relation to the conditions of detention, he argued that the account he had given to the asylum authorities very much reflected his personal experiences: he had stated that the prison had been dirty, that it had been very hot, that detainees had had lice, that there had been fixed meal times , and that he had been taken outside in the mornings and evenings to relieve himself . In light of this monotonous pattern, the mistake , if any, had been on the part of the interviewer , who had asked for a description of specific events. In that regard, the applicant argued that his account that he had been beaten with a wooden stick and kicked wh ile lying on the ground in front of everyone as punishmen t related to a specific personal event. He emphasised that the interviewer had not questioned him about his scars, which he alleged were the result of that incident, or about specific events surrounding his escape from prison, pointing out that the latter issue had been raised by the representative of the non-governmental organisation who had been present as a neutral witness at the second hearing. As to the time between his escape from prison and his departure from Eritrea, the applicant argued that his first statement that he had left Eritrea one week after escaping from prison , and his second statement that he had spent two days at his parents ’ home in A. during that time, were not contradictory .

25 . W ith regard to the circumstances of his departure from Eritrea, he emphasised that his account had been consistent as regards the time when he and the smuggler had left the village and when they had been apprehended by Ethiopian soldiers. It was only logical that he could not make detailed statements about the area, as he had no knowledge of that area and had fled during the night. His most prominent memory related to the fears he had experienced when they lost the ir way. He could also recall the exact word s in Amharic used by the Ethiopian soldiers when they had been apprehended. H is account , which revolved around feelings of thirst, hunger and fear , and which mentioned t he many thorn bushes in their way and the howling of hyenas , corresponded to his young age and poor level of education. Moreover, the State Secretariat for Migration had wrongfully concluded that there was a contradiction in his statements concerning the smuggler ’ s reaction when they had lost the ir way.

26 . Lastly , the applicant submitted that he did not belong to any of the groups of pe ople who could possibly obtain a visa to exit Eritrea. Referring to a letter from UNHCR – the original of which he presented – stating that he was registered in the Hitsas refugee camp in Ethiopia on 8 November 2013, and to the fact that crossing the border to enter Ethiopia by land was always unlawful (see paragraph 49 below ) , he argued that he had prove d his illegal exit from Eritrea.

27 . In conclusion, the applicant argued that he was a refugee as defin ed in section 3 of the Asylum Act , on account of his fear of ill-treatment for having deserted from military service. In the alternative , he claimed to qualify for temporary admission because of “ subjective post-flight grounds ” (as set out in section 54 of the Asylum Act) , notably his illegal exit from Eritrea and his asylum application in Switzerland. Further, i n the alternative , he alleged that his removal to Eritrea was neither permitted in the light of Article 3 of the Convention nor reasonable, entitling him to temporary admission to Switzerland.

28 . On 9 May 2016 the Federal Administrative Court rejected the applicant ’ s appeal , finding that he had failed to credibly demonstrate his asylum claim. It noted that i t was not apparent why the applicant ’ s age – he was 24 years old at the time of his interview s – or level of education should lead to different conclusions as to the credibility of his account , and considered that his two statements concerning the end of his schooling , either in 2005/2006 or 2007/2008, could not be reconciled with each other or with his student identity card , which indicated that he was a student in 2010. The court also considered that the applicant ’ s statements regarding the time and content of his military training lacked substance. It commented that, even if the applicant had left the military shortly after reporting for duty, he could be expected to provide a detailed and specific account of it , given that he merely had to talk about something which he had experienced in person. Viewing the duration and dates of the applicant ’ s detention as key elements of his asylum claim, the court noted that a discrepancy of eight months as regards the duration , and the different dates given in the first two hearings as regards the end of the detention , constituted fundamental contradictions which could not be resolved by his references to conditions of poor hygien e and being let outside twice a day to relieve himself.

29 . Referring to its judgment in an earlier case concerning Eritrea, the c ourt reiterated that the only way to exit Eritrea legally was with a valid passport and an additional exit visa , and that the practice concerning the issuance of an exit visa was very restrictive. They were issued to a few people who were considered loyal , in exchange for payment of significant sums. As a rule, children aged 11 or more , men under the age of 54, and women under the age of 47, were not granted exit visas. Pe ople attempting to leave the country without authorisa tion risked their life , as the border guards were under order s to prevent attempts to flee by way of targeted shots ( a “shoot to kill ” policy), in addition to imposing punishment as set out by law. The Eritrean G overnment viewed illegal exits as an indication of political opposition , and tried to get both the reduction in defen c e readiness and the mass exodus under control through draconi an measures .

30 . The Federal Administrative Court noted that finding that the applicant had concealed the true circumstances of his departure was not in itself sufficient to conclude that he had left the country legally. However , the burden of proof did not shift to the authorities , and the applicant was required to provide a substantiated and consistent account concerning the reasons for and circumstances of his departure. Considering that his account given at first - instance level was not credible, which also raised doubts about his overall credibility, and that he had not provide d comprehensible explanations in his submission s on appeal, the court found that the applicant had failed to credibly demonstrate that he had left Eritrea illegally.

31 . The Federal Administrative Court added that the applicant could not rely on the letter from UNHCR stating that he was registered in the Hitsas refugee camp in November 2013, since the conditions in Ethiopian refugee camps were chaotic and , in the case of pe ople of Eritrean origin, there was no comprehensive assess ment of whether they faced persecution at the time they were registered in those camps . This was supported by the wording of the registration , which read that it constituted a recognition prima facie that the applicant was a refugee within the mandate of UNHCR.

32 . Furthermore , the c ourt found that the appl icant ’ s removal was possible, permitted and reasonable within the meaning of section 83 ( 1 ) - ( 4 ) of the Aliens Act. In particular, the applicant was a young man in good health who had a support network in his hometown, notably his wife and their son , who lived in the same house as his parents.

33 . On 19 May 2016 the S tate Secretariat for Migration set a deadline for the applicant ’ s voluntary departure, which passed on 17 July 2016.

II. RELEVANT DOMESTIC LAW AND PRACTICE

Section 3 – Definition of the term refugee

“1 . Refugees are persons who , in their native country or last country of residence , are subject to serious disadvantages or have a well-founded fear of being exposed to such disadvantages for reasons of race, religion, nationality, membership of a particular social group , or owing to their political opinions.

2 . Serious disadvantages include a threat to life, physical integrity or freedom , as well as measures which exert intolerable psychological pressure. Motives for seeking asylum which are specific to women must be taken into account.

3 . Persons who are subject to serious disadvantages or have a well-founded fear of being exposed to such disadvantages because they have refused to perform military service or have deserted are not refugees. The provisions of the Convention of 28 July 1951 relating to the Status of Refugees are reserved.

4 . Persons who claim grounds based on their conduct following their departure which are neither an expression nor a continuation of a belief already held in their native country or country of origin are not refugees. The provisions of the Convention of Refugee Convention are reserved.”

Section 5 – Ban on refoulement

“1 . No person may be forced in any way to return to a country where his or her life, physical integrity or freedom is threatened for any of the reasons stated in section 3 ( 1 ), or where he or she would be at risk of being forced to return to such a country.

2 . The ban on refoulement may not be invoked if there are substantial grounds for the assumption that, because the person invoking it has a legally binding conviction for a particularly serious felony or misdemeanour, he or she represent s a threat to Switzerland ’ s security or is to be considered dangerous to the public.”

Section 7 – Proof of refugee status

“1 . Any person who applies for asylum must prove or at least credibly demonstrate his or her refugee status.

2 . Refugee status is credibly demonstrated if the authority regards it as proven on the balance of probabilities.

3 . In particular, c ases are not credible if they are unfounded in relation to essential points or are inherently contradictory, do not correspond to the facts , or are substantially based on forged or falsified evidence.”

Section 54 – Subjective post-flight grounds

“Refugees shall not be granted asylum if they became refugees in accordance with section 3 only by leaving their native country or country of origin , or as a result of their conduct after their departure.”

The relevant provision of the Aliens Act of 16 December 2005 ( Bundesgesetz über die Ausländerinnen und Ausländer , 142.20) provided as follows:

Section 83 – Order for temporary admission

“1 . If the enforcement of removal or expulsion is not possible, not permitted or not reasonable, the SEM [ State Secretariat for Migration] shall order temporary admission.

2 . Enforcement is not possible if the foreign national is unable to travel or be taken to his or her native country , or country of origin , or a third country.

3 . Enforcement is not permitted if Switzerland ’ s obligations under international law prevent the foreign national from making an onward journey to his or her native country, country of origin , or a third country.

4 . Enforcement may be unreasonable in respect of foreign nationals if they are specifically endangered by situations such as war, civil war, general violence and medical emergency in their native country or country of origin.

...

8 . Refugees for whom there are reasons to refus e asylum in accordance with sections 53 and 54 [of the Asyl um Act] shall be granted temporary admission. ... ”

III. RELEVANT COUNTRY INFORMATION ON ERITREA

A. Reports by the UN C ommission of I nquiry on H uman R ights in Eritrea

1. The first report ( 2015 )

36 . On 5 June 2015 the UN Commission of Inquiry on Human Rights in Eritrea (“the Commission”) presented the 483-page detailed findings of its first report to the UN Human Rights Council (A/HRC/29/CRP.1) .

37 . It observed that , pursuant to Article 5 of Proclamation No. 82/1995 (National Service Proclamation), all Eritrean nationals between the ages of 18 and 50 had a duty to participate in national service (§ 1178 of the report ). Exemptions were applied very narrowly in practice , and mainly concerned pe ople with severe permanent disabilities , as well as married women and single mothers (§§ 1193-1210). Those between the age s of 18 and 40 had a duty to participate in an 18-month period of active national service, comprising six months of military training in a training centre , followed by twelve months of active military service and development work in a combat force unit (§ 1179) . The Government extended the statutory national service period of 18 months and made it an indefinite period, effectively leading to a constant state of general mobilisation, arguing that the “no war, no peace” situation with Ethiopia justified the indefinite extension of the period of conscription (§ 1181). N ational service was also designed to contribute to the economic reconstruction of the country following the struggle for independence , and conscripts were assigned to perform a broad set of tasks, going far beyond military core functions, and work ing in the development, construction and maintenance of infrastructure projects, in the agricultural sector and in the fledgling industrial and mining sectors (§§ 1176-1177).

38 . The Commission documented a pattern of conscription into national service at an early age with no prospect of formal discharge or release for other reasons , noting that there was a clear pattern of conscription beyond the statutory period of 18 months of national service (§ 1250). Conscripts we re not informed about the length of time they we re expected to serve beyond the statutory 18 months , and had to remain in national service when the mandatory perio d of active national service lapsed (ibid.).

39 . While the majority of conscripts were sent to the Sawa military training camp, certain conscripts were sent to other military training camps (§ 1294). The latter category of conscripts included children and adults who had been caught attempting to flee and were thus seen as “traitors”; individuals who were being punished for the conduct of their parents; those who had not reach ed twelfth grade because they had dropped out of school, had not attend ed classes or had not pass ed exams; and those suspected to have purposely repeated classes several times to avoid reaching twelfth grade and being sent to Sawa (ibid.). One of the most notorious camps was opened in Wi ’ a in around 2003 and was in operation until its closure sometime before 2010 following a high number of deaths among conscripts (§ 1295). Testimonies corroborated by satellite imagery indicated that Wi ’ a ha d subsequently been reopened (ibid.). The Commission found that the harsh treatment, the lack of a clear distinction between training and labour, combined with the blurred line between trainees and detainees suggested that Wi ’ a may have been used as a re-education and correction camp rather than a normal training centre (§ 1302).

40 . Article 37 of the National Service Proclamation provided for punishments for the non-performance of military service, without prejudice to more rigorous punishment under the 1991 Transitional Penal Code of Eritrea (§ 1234). Under the Transitional Penal Code, t his statutory offence could lead to imprisonment for longer periods of time, up to life imprisonment in the case of desertion in times of emergency, general mobilisation or war (ibid.). Desertion from active service could be punished by the death penalty (ibid.). In practice, draft evaders and deserters, if caught, were severely punished (§ 1241) , and the treatment of apprehended draft evaders and deserters during detention often amounted to torture, cruel, inhumane or degrading punishment (§ 1389).

41 . The Commission further noted that Eritrea imposed severe restrictions on citizens ’ departure from the country (§ 400). Exit visas were required (§ 401). The criteria and conditions for grant ing an exit visa were not provided for by law and were left to the determination of the Government (§ 402) . Testimony col lected by the Commission revealed that exit visas we re issued to certain individuals like o lder women , individuals who had c ompleted national service , when the nature o f their occupation required regular travel , and i ndividuals who needed medical treatment abroad (§§ 403 -404) . Proclamation No. 82/1995 prohibited Eritrean citizens of military age from go ing abroad unless they could prove that they had fulfilled their national service duty or were permanently exempt from doing so (§ 406).

42 . The Commission considered that the testimony collected indicated that the border was relatively porous (§ 416). Escapees usually crossed the border on foot (§ 417) and often hire d smugglers who knew where border guards were stationed (§ 418).

43 . As regards individuals forcibly returned to Eritrea, the Commission stated:

“ 431. Individuals forcefully repatriated are inevitably considered as having left the country unlawfully, and are consequently regarded as serious offenders, but also as ‘ traitors. ’ A common pattern of treatment of returnees is their arrest upon arrival in Eritrea. They are questioned about the circumstances of their escape, whether they received help to leave the country, how the flight was funded, whether they [had] contact with opposition groups based abroad, etc. Returnees are systematically ill ‑ treated to the point of torture during the interrogation phase.

432. After interrogation, they are detained in particularly harsh conditions, often to ensure that they will not escape again. Returnees who spoke to the Commission were held in prison between eight months to three years. Male returnees from [country A] were held on Dhalak Island after a few months of detention at Adi Abeito. Deportees from other countries were held in prisons such as Prima Country and Wi ’ a.

433. Witnesses who spoke to the Commission noted the severe conditions during their detention. They were made to undertake forced labour and were frequently punished by prison guards for inconsequential matters. [Country A] returnees recounted that, on one occasion, they had been reportedly even denied drinking water where they were detained at Dhalak Island where temperatures often soared to 50 degrees Celsius. As a consequence, many fell sick after drinking unsafe water.

...

435. At no point are returnees given opportunity to contact their families, nor are they informed of t he length of their detention. Relatives find out about individuals who have been forcefully repatriated only when the latter manage to escape from the prison or the national service, or flee the country another time. After their release, women and accompanied children are usually allowed to go home. Male unaccompanied minors and those of draft age are sent to military training.

436. The Commission found however two exceptions to the rule that returnees are arrested, detained and forced to enlist in the national service upon their arrival in Eritrea. A group of Eritreans was returned from [country D] with a letter certifying that they had paid the 2 per cent Rehabilitation Tax and had already been detained several years in [country D]. The witness had himself been imprisoned for three years in [country D]. He was given a permit to return to his hometown, but which had to be renewed every two months. He left Eritrea again shortly aft er being deported. The other case concerned forced repatriations to Eritrea in 2014, where seven older men were reportedly freed while the younger men who were returned in Eritrea at the same time were not released.”

2. The second report ( 2016 )

44 . On 9 May 2016 the Commission presented its second report to the UN Human Rights Council (A/HRC/32/47), which stated:

“ 33. On 8 April 2015, Yemane Gebreab, at the Bruno Kreisky Forum for International Dialogue, announced that Eritrea intended to limit its military/national service programmes to 18 months. Eight months later, however, the Government stated:

Eritrea has no option but to take necessary measures of self-defense that are proportionate to the threat it faces ... This is the reason why National Service – limited by law to 18 months – remains prolonged.

34. In February 2016, the Minister for Information, Yemane Ghebremeskel, confirmed that there were no plans to limit military service programmes, stating that ‘ demobilization is predicated on removal of the main threat ’ , and ‘ You are talking about prolongation of national service in response to ... continued belligerence by Ethiopia. ’

35. The commission emphasizes that mandatory military/national service is not necessarily a human rights violation. What distinguishes the military/national service programme in Eritrea from those in other States is (a) its open-ended and arbitrary duration, which routinely exceeds the 18 months provided for in a decree issued in 1995, frequently by more than a decade; (b) the use of conscripts as forced labour in a wide range of economic activities, including private enterprises; and (c) the rape and torture perpetrated in military camps, and other conditions that are often inhumane.

...

37. In its first report, the commission reported extensively on cases of arbitrary detention, enforced disappearance, and torture and other cruel, inhuman or degrading treatment in detention centres, military and civilian, official and unofficial.

38. The commission interviewed many Eritreans who had fled the country in the previous two years and reported that the violations described continue. Almost all of those arrested are detained in violation of fundamental rules of international law. Apart from those accused of minor common crimes or misdemeanours, most are detained without any form of judicial proceeding whatsoever. In the vast majority of those cases, the families of those detained receive no official information about the fate of their relatives. Lastly, many of those detained who spoke with the commission – either because they had been released or because they had escaped – described various forms of torture inflicted on them to obtain information or to punish them for alleged wrongs, or simply to create a general climate of fear.

...

57. In its resolution 29/18, the Human Rights Council extended the mandate of the commission of inquiry to investigate systematic, widespread and gross violations of human rights in Eritrea, including where these violations may amount to crimes against humanity.

...

65. ... [T]he victims of the military/national service schemes in Eritrea are not bought and sold on an open market. Rather, the powers attaching to the right of ownership in Eritrea are revealed by (a) the uncertain legal basis for the national service programmes; (b) the arbitrary and open-ended duration of conscription, routinely for years beyond the 18 months provided for by the decree of 1995; (c) the involuntary nature of service beyond the 18 months provided for by law; (d) the use of forced labour, including domestic servitude, to benefit private, PFDJ-controlled and State-owned interests; (e) the limitations on freedom of movement; (f) the inhumane conditions, and the use of torture and sexual violence; (g) extreme coercive measures to deter escape; (h) punishment for alleged attempts to desert military service, without an administrative or judicial proceeding; (i) the limitations on all forms of religious observance; and (j) the catastrophic impact of lengthy conscription and conditions on freedom of religion, choice, association and family life.

...

68. The commission concludes that there are reasonable grounds to believe that, within the context of military/national service programmes, Eritrean officials exercise powers attaching to the right of ownership over Eritrean citizens. It also determines that, despite the justifications for a military/national service programme advanced in 1995, the programmes today serve primarily to boost economic development, to profit State-endorsed enterprises and to maintain control over the Eritrean population in a manner inconsistent with international law. The commission therefore finds that there are reasonable grounds to believe that Eritrean officials have committed the crime of enslavement, a crime against humanity, in a persistent, widespread and systematic manner since no later than 2002.”

45 . These aspects are further elaborated on in the detailed findings which the Commission presented to t he UN Human Rights Council on 8 June 2016 ( A/HRC/32/CRP.1 ).

B. Report s by EASO

1. The November 2016 report “Eritrea – National Service and Illegal Exit”

46 . In November 2016 EASO published a country of origin information report entitled “Eritrea – National Service and Illegal Exit”. The report was authored by the Swiss State Secretariat for Migration , which also published the German version of the report in June 2016, and reviewed by the Norwegian and Swedish asylum and migration departments prior to its publication. It constituted a partial update of the May 2015 EASO report “Eritrea Country Focus”.

47 . In its introductory chapter (pp. 13-17), t he November 2016 report elaborated on the methodology employed and the sources used. A fact ‑ finding mission to Eritrea and neighbouring countries had been conducted in February and March 2016 . T he report stated:

“[It] was not possible to visit any detention facilities during official visits made to Eritrea. Other migration services, international organisations and diplomats likewise had no access to Eritrean prisons. ...

Access to information about Eritrea, particularly on human rights issues, is difficult. The Eritrean Ministry of Information controls all media in Eritrea. Researchers, journalists and representatives of human rights organisations are generally unable to conduct research in the country, or can do so only to a very limited degree. The Eritrean authorities publish very little detailed information about the national service. There is a similar lack of transparency regarding the implementation of legislation on national service and illegal exit; and the authorities do not publish any guidelines or implementing provisions. This means that there is a lack of essential sources of information on those topics that are relevant to international protection status determination. The information available is based almost exclusively on the following three categories of sources:

Statements made by the Eritrean government : The Eritrean government generally dismisses all allegations of human rights violations. It makes its statements public, inter alia , via Eritrea ’ s state media, including the portal www.shabait.com. Representatives of the Eritrean government, of the ruling party (the People ’ s Front for Democracy and Justice, PFDJ) and of organisations close to the government also constantly make statements to foreign media and delegations. These statements can be accessed via the media and in various Country of Origin Information (COI) reports.

Assessments by persons in Eritrea : Residents of Eritrea (Eritreans and foreign nationals) are best placed to report on the current situation in the country. However, in the experience of the SEM, their knowledge is limited: just like foreign visitors, people from Asmara who are normally contacted by journalists and other observers have no access, for example, to prisons or military camps, meaning that information they provide is based on reports from acquaintances. In recent years, the Eritrean authorities have appeared to tolerate greater criticism. However, residents of Eritrea are rather cautious and reserved when speaking to foreigners (in particular foreign media and official delegations). Accordingly, situation reports provided by Eritrean residents and foreign observers (diplomats, employees of international organisations) tend to be more positive than those provided by exiled Eritreans. These assessments – predominantly assessments by foreign observers – are accessible, inter alia, in the reports produced by various national COI units in Europe.

Assessments by persons who have left Eritrea : Reports by human rights organisations in particular are based to a considerable extent on statements by people who have left Eritrea. In the experience of the SEM, these organisations are mainly contacted by people who claim to have had terrible experiences before leaving the country or who, for other reasons, would like to draw attention to abuses in Eritrea. The Eritrean government is likewise frequently the subject of harsh criticism in pro ‑ opposition diaspora media; there are, however, also pro-government diaspora media. The views of people who have left Eritrea can mostly be found in human rights reports produced by organisations such as Human Rights Watch and Amnesty International or by the US Department of State, and also sometimes in the media in the destination countries of Eritrean migrants.

...

The Swiss [Country of Origin Information] unit ’ s views on the sources used in the individual sections are as follows:

Legal position : The Eritrean legal provisions on national service and illegal exits are publicly accessible. However, the authorities ’ and the military ’ s internal guidelines are not accessible despite apparently also being applied in those areas.

Position of the Eritrean government : This information is mostly based on statements made by representatives of the Eritrean government, of the authorities, of the ruling party (the PFDJ) or of one of two organisations that are close to the government, the National Union of Eritrean Youth and Students (NUEYS) and the National Union of Eritrean Women (NUEW). During the fact-finding mission, these representatives provided information on the topics covered in this report. The Eritrean authorities proofread and confirmed all the statements taken from the discussions held and used in the report. In addition to the interviews conducted as part of the fact ‑ finding mission, public statements made by representatives of the Eritrean government were also included in the report. The government representatives provided information about the arrangements for dealing with deserters, draft evaders etc. that is at odds with the legal position. However, neither they nor other people interviewed as part of the fact-finding mission were able to substantiate the information provided by means of guidelines, court judgments or statistics. Nor was it possible to discuss issues relating to national service with the competent ministry – the Ministry of Defence.

Views of international observers in Asmara : The assessments of the situation by people in Eritrea were mostly obtained by interviewing representatives of international organisations and foreign embassies as well as a number of other residents of Asmara. The views of the international observers are based almost exclusively on anecdotal knowledge acquired from conversations with Eritreans and on the conclusions that they themselves have drawn from that information. Since these observers are based in Asmara, this anecdotal knowledge relates mostly to the capital and its surrounding areas; regional variations cannot be ruled out. They also do not have access to guidelines, court judgments or statistics about the arrangements for dealing with deserters, draft evaders etc. Since only relatively few international representatives reside in Asmara and they are in frequent contact with one another, there is a risk of information round-tripping and false corroboration. In addition, in isolated cases, this section also provides information obtained from interviews with Eritreans living in Asmara. At the request of the people interviewed for this report, the sources of all information of this kind have been made anonymous.

2015 and 2016 reports : Four different categories of reports were used:

• Detailed human rights reports published by the UN Commission of Inquiry in June 2015 and 2016 and by Amnesty International in December 2015. Both sets of reports are mainly based on statements made by people who have left Eritrea for reasons that make them very critical on the situation there, and in addition on external experts and public sources. The statements used by Amnesty Internatio nal relate to 2014 and 2015 , and those used by the UN Commission of Inquiry cover the period from 1991 to 2015/2016.

• Annual summary reports on the human rights situation in 2015 produced by Amnesty International, Human Rights Watch and the US State Department. These reports are based on a variety of sources, primarily from outside Eritrea, which in most cases are not explicitly stated.

• Reports by Country of Origin Information (COI) units in the United Kingdom, Norway and Sweden that were published following fact-finding missions conducted in late 2015 or in the spring of 2016. In producing these reports, and in addition to the findings of the fact-finding missions, these units (with the exception of the UK unit) also used findings taken from other reports and the views of experts from outside Eritrea.

• Various relevant media reports from Europe, Eritrea and the Eritrean diaspora that are based on a very wide variety of sources. ”

48 . In substance, the report stated:

“According to most sources consulted for this report, deserters apprehended within Eritrea are usually returned to their military unit or civilian duty and punished. These punishments are imposed extrajudicially by their superiors. There is no possibility of appeal. However, the treatment of deserters appears to have become less harsh in recent years. Most sources report that first time offenders are now usually detained for several months. Punishment for deserters from the military part of national service is reportedly more severe than punishment imposed on those deployed in the civilian part. As deserters are not tracked down systematically, a number of them effectively go unpunished.

...

According to almost all sources, individuals who leave Eritrea illegally are also subjected to extrajudicial punishment. It is unclear who is in charge of imposing penalties. No judgments are made public and there is no possibility of appeal. However, the policy currently applied by the authorities appears to allow for shorter prison sentences than those enshrined in the law. According to most reports, the detention period now commonly lasts a few months up to two years, depending on the circumstances. After being released, deserters have to resume their national service, while draft evaders are conscripted for military training. The alleged ‘ shoot-to-kill order ’ at the border is not followed strictly, according to most consulted sources. However, shootings may occur. For voluntary returnees from abroad who had previously evaded draft, deserted or left the country illegally, the draconian laws are reportedly not applied at the moment, provided they have regularised their relationship with the Eritrean authorities prior to their return. According to a new, unpublished directive, such returnees are exempt from punishment. It is understood that the majority of the individuals who have returned according to this directive have effectively not been persecuted. Nonetheless, concerns remain.

There is no legal certainty, because the directive has never been made public. Furthermore prospective returnees are obliged to pay a diaspora tax (2% tax) to an Eritrean representation abroad and to sign a ‘ letter of regret ’ in case they have not yet fulfilled their national service duty. It should also be noted that not all Eritreans are able to return this way. For example, persons who were critical of the Eritrean government during their time abroad are either denied return or would risk detention upon their return. So far, the majority of Eritreans who returned did so voluntarily and only temporarily. The long-term consequences of returns on a permanent base are still unknown.

There is hardly any information available regarding the treatment of forcibly returned persons. In the last few years, only Sudan (and possibly Egypt) forcibly repatriated Eritreans. As opposed to voluntary returnees, those forcibly returned are not able to regularise their relation with the Eritrean authorities prior to returning. The few available reports indicate that the authorities treat them similarly as persons apprehended within Eritrea or while leaving illegally. For deserters and draft evaders, this means being sent back to national service after several months of detention. Regularisation is not necessary for persons who have not reached conscription age yet or who have fulfilled their national service duty already. Nevertheless, it cannot be excluded that adults are punished for non-payment of the diaspora tax or for illegal exit.

Over the last few years, the Eritrean authorities have announced several reforms of the national service. Most notably, they promised to limit the length of duty to 18 m onths starting from the 27 th conscription round. This has not been fulfilled yet. National service remains open-ended and conscription lasts for several years. According to sources consulted, a growing number of conscripts who had been deployed in civilian roles are discharged once they h ave served for between 5 and 10 years. However, no reliable information is available on the demobilisation and dismissal of conscripts assigned to the military part of national service. In early 2016, the authorities announced a pay rise in the civilian part of the national service. According to sources consulted, implementation has already started. ”

2. The May 2015 report “Eritrea – Country Focus”

49 . The report set out, inter alia , the following:

“ Education

... In theory, school attendance is compulsor y for the first eight years . At the end of the eighth year, children sit a national examination before moving on to secondary school; the pass rate stands at about two-thirds .

...

Exemptions from national/military service

... In the past, religious leaders of the four official religious communities ... were partially exempt from military service but this stopped being the case in 2010 at the latest.

...

Legal exits

... [Legal] [b]order crossings are only possible at one of the allocated border control points, which include Asmara airport, Massawa sea port and several border crossings into Sudan ... ”

3. The report “Latest asylum trends – 2016 overview”

50 . The report set out, inter alia , that in 2016 92 % of the decisions on applications for international protection from Eritrean nationals taken in Member States of the European Union plus Switzerland and Norway were positive , that is resulting in the grant of either refugee status or another form of protection .

C. The 2009 Country Report by the US Department of State

51 . The 2009 Country Report on Human Rights Practices – Eritrea, published by the US Department of State on 11 March 2010, stated, inter alia :

“Students who received poor grades in high school had in the past been sent to the Wi ’ a Military Camp in lieu of being allowed to complete the academic year.”

D . Relevant case-law of the Upper Tribunal (Immigration and Asylum Chamber) of the United Kingdom

52 . On 10 October 2016 the Upper Tribunal (Immigration and Asylum Chamber) of the United Kingdom (hereafter “the Upper Tribunal) issued country guidance in its judgment in the case of MST and Others (national service – risk categories), Eritrea CG [2016] UKUT 00 443 (IAC), which contained a comprehensive examination of all up-to-date material on Eritrea. The judgment ’ s findings included the following:

“ ... National service

...

280. [T] he preponderance of the evidence points strongly to the conclusion that the Eritrean regime of military/national service (excluding civilian national service and the people ’ s militia), is characterised by a system that often responds to transgressors with harsh and disproportionate punishments. We exclude from this conclusion civilian national service and the people ’ s militia because by contrast the evidence does not demonstrate that punishment for transgressions by persons evading or deserting from one or the other is either as likely or as severe in nature.

281. We would accept that the preponderance of evidence also indicates that roundups (giffas) are happening les s frequently and that the ‘ shoot to kill ’ policy is now intermittent and arbitrarily applied and that punishment of family members or associates may not be as common as it was, but these are only some of the regime ’ s repertoire of punishments, and there is a substantial body of evidence, including the US State Department reports, indicating that the generality of evaders and deserters are harshly punished and this is a common thread running through the majority of source evidence. We note that the 2015 [UN Commission of Inquiry] Report ... refers to the grant of an amnesty to deserters in November 2014, but this was from detention and the Report does not suggest this represented a change of government policy. The main evidence concerning this matter on which the respondent relies is that from Eritrean government ministers and interviews with individuals during the [ f act- f inding m ission of the United Kingdom] and we have explained why we feel that this evidence should be approached with caution ... We have taken into account the evidence of [Amnesty International] ... that punishment for deserters is generally more severe although this is arbitrary and that the generality of evaders and deserters are punished with imprisonment for varying periods. Those caught on the border trying to flee are almost always subjected to periods of arbitrary detention. Generally ... those arrested for evading service are detained for some time between one and six months. The reports demonstrate ... a high level of variation which is said to be indicative of the arbitrary nature of punishments that are at the discretion of officers. The EASO Report [of May 2015] concludes ... that deserters and evaders are punished by imprisonment if caught within the country before being able to leave or on return at the airport and that punishment is harsh being more severe for deserters. [The] evidence [of expert witness Professor Kibreab] throughout is that deserters/evaders will be subject to persecution.

282. The 2015 [UN Commission of Inquiry] Report ... reports arbitrary detention, enforced disappearance, torture and mistreatment generally in Eritrean detention centres. The Commission spoke to those who had fled in the past two years and reported that they had been subject to ill-treatment and detained without due process. The Commission ... reports arbitrary detention for periods ranging from months to years, enforced disappearances ... and torture ... EASO reports ... poor conditions in detention. The Swiss fact-finding report of March 2016, to which several references are made in the new Home Office [Country Information and Guidance] of August 2016, considers that even though the treatment of deserters appears to have beco me less harsh in recent years, ‘ [m]ost sources report that first time offenders are now usually detained for several months ’ ... Given that we consider anything beyond very short-term detention in Eritrea to create a real risk of ill-treatment, this confirms our view that deserters/evaders continue to face a real risk of persecution.

283. To summarise, we reject the respondent ’ s case that enforcement and punishment is reserved for those who are involved in oppositional activity over and above desertion or evasion. It is impossible in our view to derive from the evidence as a whole any other conclusion than that for Eritreans inside the country any evasion of military service or desertion still carries a real risk that the generality of transgressors will be subject to treatment which amounts to persecution as well as serious harm.

...

287. As regards the eligibility requirements for national service, age (and duration) in particular, we will deal below with the age requirements when considering the categories of lawful exit visas ... But in a nutshell we consider that t he age limits for national service are likely to remain the same as stated in MO [(illegal exit – risk on return), Eritrea CG [2011] UKUT 00190 (IAC)] , namely 54 for men and 47 for women except that for children the limit is now likely to be 5 save for adolescents in the context of family reunification. For the people ’ s militia, the age limits are likely to be 60 for women and 70 for men.

288. In relation to duration, it is agreed on all sides that national service is indefinite and open-ended, but there is disagreement as to whether this means that it results in most Eritreans performing military/national service duties permanently or for very prolonged periods. As noted above, the respondent ’ s position is that actual performance of military/national service is variable and uncertain, but that there is a real prospect of discharge. This is in stark contrast to the position of the appellants, UNHCR and [Professor Kibreab] .

289. We accept that there are no clear statistics relating to the number of individuals in national service, but it is reasonable to infer from what evidence there is, that at any one time most people are not engaged in the performance of military or national service duties. Most sources estimate Eritrea to have a population of over 6.3 million. The 2015 [UN Commission of Inquiry] Report states that there are 201,750 active members of the armed forces, the majority being national service conscripts ... The EASO Report [of May 2015] ... states that there is no official data available regarding the number of people engaged in national service but various estimates place the figure at between 200,000 and 600,000 in recent years, approximately half of whom are assigned to active military service. [Professor Kibreab] ’ s opinion is that 9.2 per cent of the population has been conscripted over the past 20 years (the figures, he states, do not take into account those who have fled the country). The respondent does not accept [Professor Kibreab] ’ s percentage figure, claiming that it is far less. However, even if we accept [Professor Kibreab] ’ s opinion on the issue, which is the most favourable to the appellants, the figures are significant. The only logical conclusion we can draw from them is that active performance of national service duties cannot be as extensive as the appellants and UNHCR assert. The system remains indefinite and open-ended in the sense that all persons of or approaching eligible draft age or within the age limits for the people ’ s militia remain obliged to perform military/national service; but it is a distinct matter whether persons have to actually perform military/national service and for what periods of time. We shall return to the possible implications of this conclusion when we deal further with demobilisations/discharges ...

290. We also consider the evidence to indicate that discharge/release is a more common phenomenon than the appellants contend. We will address this issue more fully when we deal with demobilisations/discharges and with draft evaders and deserters ...

294. We note that there is wide recognition that (separate from the legal possibilities for exemption, which all agree are limited by legislation to medical cases), a significant number of people appear able to obtain exemptions based on contacts and/or bribes. We take the principal thrust of the evidence regarding such avenues as being that national service is not necessarily an unavoidable exp erience for everyone in Eritrea.

...

304. A person starts national service at age 18 or indeed even younger in some cases. It is very unlikely that a conscript will be released within the first 18 months of service when a conscript is engaged in active national service (which comprises six months military training and 12 months military service). Our understanding is that immediately after this period conscripts are redeployed. The evidence points strongly, therefore, to a system which conscripts young people at 18 (or earlier) and then requires them to continue national service uninterrupted beyond completion of the initial 18 months. When a person starts national service, the term they will be expected to complete is not known and to this extent it is arbitrary and indefinite. Ordinarily, by the time they are in their mid-20s (unless they have been discharged or dismissed or released) they are likely to have been in national service for 7 years. The critical issue is how long the period is likely to be for them to be accepted to have completed national service in the eyes of the Eritrean authorities. Here there is evidence going both ways.

...

306. We are bound to say we have had very considerable difficulty deciding this issue, notwithstanding the preponderance of sources that describe national service as protracted, for two reasons. First, because for reasons set out earlier we consider it likely that release is commonplace. Secondly because (as also noted earlier) the figures of persons involved in national service at any one time appear to indicate that 9 out of 10 persons are not engaged in national service duties. If we had felt able to draw inferences from these two findings alone, we might well have concluded that the Eritrean authorities are likely to regard 7 years as being long enough for them to be satisfied an Eritrean citizen has completed national service. We are certainly satisfied that the great majority of Eritreans begin national service at the age of 18 (if not earlier) and continue in national service beyond the 18 months period and that this means that ordinarily, by the time they reached 25 (if they have not been discharged, dismissed or released), they would have performed 7 years of national service. As a corollary, we would have concluded that the category of those who have left Eritrea illegally who would be perceived on return as draft evaders or deserters would be confined to those who were under the age of 25 or could otherwise show that they had not yet served 7 years. However, we do not think inferences can be drawn from these two findings alone. It seems to us that the broader body of evidence identifying national service as prolonged must be weighed in the balance and accorded due weight. Even in relation to the evidence regarding release, it is likely that in a significant number of cases release is simply de facto , without it being confirmed by official documentation which makes it likely that it would be difficult for the generality of beneficiaries to show that their national service was formally complete.

...

Eligibi lity f or national service and exit visas

308. By Article 17 of Proclamation No.82/1995 an Eritrean citizen ‘ under the obligation of national service ... may be allowed to travel abroad ’ by producing evidence that he or she is exempted or has completed his or her service or by producing a registration card and entering into a security bond. Lawful exit from Eritrea requires an exit visa issued by the Department of Immigration. According to the 2015 [UN Commission of Inquiry] Repo rt ... exit visas are issued to certain individuals without difficulty and in this regard mention is made of three categories: older women; individuals who have completed national service when the nature of their occupation requires regular travel; and conscripts travelling for official business for the government, although it is emphasised that the system operates arbitrarily.

...

326. Of course, in regard to all ... categories we accept there are continuing uncertainties and contradictions (as highlighted by the EASO Report [of May 2015] ) and a certain degree of arbitrariness (as highlighted by [Professor Kibreab] and the [UN Commission of Inquiry] Reports). [The] categories represent therefore only those mostly likely to be available; there remains the possibility in any individual case of denial.

...

328. We conclude that the categories of lawful exit have not significantly changed since MO . The Eritrean system of exit visas continues to afford, and to be perceived by a significant number of Eritreans as affording, real, albeit restricted, possibilities for them to avail themselves of and accordingly we would list the exit categories as follows (where the categories are different from those given by EASO [in May 2015] , they are underlined):

...

The 2 per cent tax and the regret letter

...

333. The weight of the evidence points very much in the direction that the letter and the tax do not guarantee safety for Eritreans returning; rather they enable them to access consular services. There is scant evidence of anyone who has not been naturalised in another country paying the tax and/or signing the letter and returning safely or otherwise. We accept [Professor Kibreab] ’ s evidence about this, which was very much corroborated by evidence from other sources. There being insufficient detail about the returnees to draw conclusions, we would have reached this conclusion independently in any event. Apart from the two exceptions referred to by the [UN Commission of Inquiry] , it would appear that the bulk of the examples cited concern or may concern persons who voluntarily returned, which in our view (as set out below when dealing with failed asylum seekers and forcible returns ... ) puts them in a different category.

334. Suffice to say for the purpose of this section, that we do not accept that the evidence goes anywhere close to establishing that the payment of the tax and the signing of the letter would enable draft evaders and deserters to reconcile with the Eritrean authorities. In relation to the letter of regret, we also have serious doubts that it can properly be described as a basis for reconciliation, since its terms amount to a confession of guilt by the person who signs it to what the Eritrean regime considers ‘ appropriate punishment ’ in the context of a regime with a very poor human rights record.

Failed Asylum Seekers

335. In MO the Tribunal ... held that failed asylum seekers as such are not at risk of persecution on return. We do not detect any enthusiasm from any of the parties for a different view being taken today. Indeed the appellants ’ expert witness [Professor Kibreab] , was adamant that failed asylum seeking could not be enough on its own to engender risk because of the main reasons highlighted in MA [(Draft evaders – illegal departures – risk) Eritrea CG [2007] UKAIT 00059] and MO that the Eritrean authorities have a vested interest in embedding abroad people who claim asylum but are in reality well-regarded by the government and that a significant number appear to be in reality supporters of the Eritrean government or able to demonstrate that they are through attendance at rallies etc.

336. We note that references can be found in some of the sources taking a different view, but here we regard the way the matter was put by the April 2015 Landinfo Report, that there was ‘ no empirical evidence ’ to support the contention that an application for asylum will lead to adverse reactions from the Eritrean authorities, as being entirely fair.

337. To the extent that any inferences can be drawn from the evidence overall, it seems to us that there is likely to be a further reason presently why the Eritrean authorities would not view the mere fact of being a failed asylum seeker adversely. This is that the Eritrean authorities consciously recognise the economic value to them of having a sizeable diaspora who send remittances and some of whose members also pay the 2 per cent tax. Rightly or wrongly, they clearly consider that many of the Eritreans who have left have done so out of a desire for economic betterment rather than asylum yet go on to claim asylum as a way of residing elsewhere. That may be a factor that has played a part in Eritrean government thinking for some time, but recent evidence does underscore how greatly the Eritrean government depends on foreign remittances. According to Crisis Group Afric a Briefing No 100 August 2014 ( ‘ Eritrea: Ending the Exodus? ’ ) remittances inject hard currency into the country ’ s meagre foreign exchange reserves, whilst bolstering the economic resilience of the families left behind and the government has become increasingly dependent on Eritreans abroad as a source of capital. It was estimated that approximately one third of Eritrea ’ s 2005 GDP came from remittances and this may have increased. Whilst there are still references in some sources to the Eritrean authorities viewing failed asylum seekers as traitors, we continue to follow MO in considering this as something only likely to be acted on in any way when there is a particular symbolic importance for Eritrea public policy e.g. when dealing with collective expulsions back to Eritrea. This last observation, however, is we think of greater importance than previously, because what we have to consider is not just how failed asylum seeking as such would be perceived, but how the Eritrean authorities would react to persons perceived as draft evaders or deserters when forcibly returned.

Illegal exit by those perceived on return to be draft-evaders or deserters

...

344. As regards the issue of how decision-makers should decide whether a person has left illegally, we see no reason to differ from the precise terms of the guidance in MO ... :

‘ (iii) ... The general position as regards illegal exit remains as expressed in MA , namely that illegal exit by a person of or approaching draft age and not medically unfit cannot be assumed if they have been found to be wholly incredible. However, if such a person is found to have left Eritrea on or after August/September 2008, it may be that inferences can be drawn from their health history or level of education or their skills profile as to whether legal exit on their part was feasible, provided that such inference can be drawn in the light of adverse credibility findings. ’

None of the parties has pointed to any evidence indicating the need for a different approach on this issue. We would next reiterate that it is incorrect of the March and September 2015 and August 2016 [Country Information and Guidance] to portray (as they certainly do in places) the position set out in MO as being that Eritreans who left illegally are considered to be, per se , at risk. The MO position was explicitly stated as being subject to three exceptions ... Indeed, UK country guidance has never asserted that the fact of illegal exit from Eritrea is of itself enough to place a person at risk.

345. ... [O] ur view is that the totality of the evidence continues to support the view that the fact of illegal exit is not of itself enough to place an individual at risk.

346. The question is, therefore, what further characteristics are needed to place a person at real risk of persecution or serious harm on return.

347. We consider two further characteristics are needed: (i) that they will be perceived on return as evaders/deserters; and (ii) that they will be persons subject to forcible return. Even then, however, we continue to think that this category is subject to certain exceptions and that they are exactly the same as those identified in MO , namely (1) persons whom the regime ’ s military and political leadership perceives as having given them valuable service (either in Eritrea or abroad); (2) persons who are trusted family members of, or are themselves part of, the regime ’ s military or political leadership. A further possible exception, requiring a more case specific analysis is (3) persons (and their children born afterwards) who fled (what later became the territory of) Eritrea during the War of Independence. We do not accept the position identified in the latest version of the Home Office [Country Information and Guidance] on Illegal Exit published on 4 August 2016 that the scope of these exceptions has widened.

...

354. What, then, is the basis for considering that those who left illegally and will be perceived on return as draft evaders/deserters would be at risk? There is first of all, the evidence as to what happens to evaders/deserters within Eritrea. As explained ... , we are satisfied that despite a lessening in the frequency of round-ups (giffas) and ‘ shoot to kill ’ operations and punishment of relatives, the treatment such persons are likely to face amounts to persecution or serious harm, since it continues to take the form of widespread recourse to detention. [The respondent ’ s counsel] conceded during the hearing that anything more than a very short period in detention in Eritrea would carry a real risk of ill-treatment and on the available evidence there is in our judgement a real risk that draft evaders/deserters regularly face more than very short-term detention. There is some evidence that some persons may, instead of detention, face assignment to military/national service, but for an initial period of time, it is likely this will be assignment to military duties and, in any event, as will be explained below, we consider that a requirement to perform national service duties, military or civilian, would constitute forced labour contrary to Article 4 of the ECHR, if not also Article 3.

355. Second, argument that the Eritrean authorities would treat returning evaders/ deserters differently from in-country evaders/ deserters seems to us insufficiently made out. Indeed, one of the most recent sources cited ... [in] the August [Country Information and Guidance] on Illegal Exit (the Swiss Report of March 2016) states that ‘ [t]he few available reports indicate that the authorities treat them similarly as pers ons apprehended within Eritrea. ’ This brings us to the second characteristic which we consider is required to bring a person within a risk category.

356. The specific category of persons with whom we are concerned are not draft evaders or deserters who have left illegally and would be making a voluntary return. In relation to the latter there are some possible examples in the evidence which suggest they can reach reconciliation with the Eritrean authorities. We have taken particular note in this regard of the sources relating to voluntary returns cited by the latest version of the Home Office [Country Information and Guidance] on Illegal Exit ... Those with whom we are concerned are persons who are or will be perceived as evaders/ deserters and who will be known to be persons who are the subject of a forcible return. Whilst we do not necessarily think the Eritrean authorities would react in precisely the same way to individual forced returnees as they have in the past to mass forcible returnees, we consider it reasonably likely that they would feel similarly impelled to adopt a punitive stance in a way they have not sometimes done to voluntary returnees. On the totality of the evidence we consider this is a reasonably likely state of affairs. We must analyse the issue of forcible returns in more depth in the next subsection.

Forcible Returns

...

366. [T]he recent evidence of forcible returns made from non-Western countries, chiefly the overland repatriations from Sudan, is really the only type of evidence we have against which to assess risk on return from Western countries ... And it constitutes evidence showing that in the last few years those who are likely to be perceived on return as draft evaders/deserters and who have been the subject of such forcible returns have met with, or are likely to have met with, ill-treatment on return. Further, recommencement of forcible returns from Europe would very likely in our judgement be seen by the Eritrean authorities as requiring them to adopt a punitive stance even in relation to persons in the aforementioned category who are returned individually. We infer that their reaction to such a re-commencement would be a matter of high importance to the regime.

367. It is possible to conjecture that the Eritrean government would feel the need, especially in the light of recent EU funding, to demonstrate a more relaxed or softer policy, such as was mooted in the [Danish Fact-Finding Mission] Report mainly (it seems) by reference to voluntary returnees. On the other hand, the evidence points more strongly to the policy imperatives of the current Eritrean government being driven not by concerns about its image in the eyes of Europe and the West but by domestic concerns about the maintenance of control and regulation of their own population and the need to show that those perceived as draft evaders or deserters would not receive preferential treatment on return. In our judgement there is a real risk that the likely reaction would therefore be similar to that given to those forcibly repatriated from Sudan and the evidence we have about that indicates such persons are likely to face treatment contrary to basic human rights.

Draft Evaders and Deserters

368. To this point our assessment of the issue of risk on return to those who left illegally and are likely to be perceived on return as draft evaders and deserters is not markedly different from MO . We now have to consider whether it remains sufficient that such persons have exited illegally and are of or approaching eligible draft age (unless falling within one of three specified exceptions).

...

370. ... [W]e conclude that the preponderance of the evidence continues to support the MO position and that, although it is reasonably likely that persons who have been released will have documentation which will enable them to travel within Eritrea, the fact that they are reservists (a term we use here simply to identify those who have been discharged/released) would not entitle them to an exit visa. Whilst release is commonplace, it appears that it is often de facto and that those who benefit would not ordinarily be given or hold official documents confirming that they have completed national service. We consider that recall is not common but that the Eritrean system operates to ensure that the great majority of those of or approaching draft age are regarded as still ‘ on the books ’ and as not having completed national service. What was noted in the EASO Report [of May 2015] regarding civilian national service and those in ministries strikes us as very pertinent: ‘ [m]any employees of ministries do not know whether they are still engaged in national service or have been dismissed ’ . We remind ourselves that the great majority of sources, including the very recent UNCOI Reports, consider that the duration of national service is prolonged. From the evidence we conclude that a person who exits Eritrea illegally and is of or approaching draft age, is likely on return to be perceived as an evader or deserter because of non ‑ completion of national service.

National service as slavery or servitude or forced labour

...

402. The principal basis on which the appellants contend that the Eritrean national service system amounts to slavery is the conclusions of the 2016 [UN Commission of Inquiry] Report to this effect. We would note that we think they are entirely right to focus on the 2016 Report ...

403. We have considerable reservations about the reasoning adopted in the 2016 Report as regards slavery and servitude.

...

405. Whilst it has oppressive features, we do not consider that the Eritrean system of military/national service constitutes anything comparable to the paradigm identified in Siliadin of ‘ the obligation for the ‘ serf ’ to live on another person ’ s property and the impossib ility of altering his condition ’ , certainly not in the context of assessing the military/national service system as a whole, whose con ditions are extremely variable ... Even those who are required to perform lengthy national service cannot sensibly be described as being compelled to live permanently on government property and whilst the possibilities for exemption or de facto demobilisation are limited, it cannot be said that there is an impossibility to alter one ’ s condition. Nor do we consider that the obligation to perform military/national service can sensibly be described as amounting to the ‘ exercise [by the Eritrean state] of a genuine right of legal ownership ... . reducing those called up to the status of an ‘ object ’ . Eritrean law does not create such a legal ownership.

406. ... However, even on the Commission of Inquiry ’ s own application of ... indicia [ of ownership ] to the Eritrean context, we do not follow how it progresses from its argument that there are certain aspects of the Eritrean system of military/national service that constitute the crime of enslavement to its conclusion that the programme generally, including civilian national service and service in the people ’ s militia, constitutes such a crime.

407. Of the ten indicia relied on to justify the finding that the system amounts to enslavement, there are at least three that can only be applied to civilian national service and the people ’ s militia with considerable difficulty: e.g. ‘ (vi) inhumane conditions ’ , ‘ (vii) torture and killing ’ (where all the examples cited relate to military national service, not civilian national service) and ‘ (x) impact on family life ’ . ...

414. We consider that very similar difficulties apply when one turns to consider whether, even if not slavery, the Eritrean system of milita ry/national service amounts to ‘ servitude ’ contrary to Article 4(1).

...

416. That leaves the issue of whether the system amounts to ‘ forced or compulsory labour ’ .

417. In this context and in light of the legal framework summarised earlier, it seems to us that the evidence we have before us is on a different footing. For one thing we have the ILO analysis and (unlike the international criminal law framework) the ECtHR has seen the ILO framework to have a bearing on interpretation of Article 4 ... For another, the ILO analysis, taken together with other sources, constitutes a considerable body of very specific evidence tending to show that the workings of the Eritrean system cannot be seen to fall under any of the exclusions set out in Article 4(3). That is important because in the course of various ILO proceedings the Eritrean government has not disputed that their military/national service system amounts to forced or compulsory labour. Their argument is directed only to their system falling under one or more of the permitted exemptions or exclusions.

418. We take first the exclusion of ‘ any service of a military character ’ (Article 4(3)(b)).

419. Paragraph 3(b) of the Article exclu des from the ambit of the term ‘ forced or compulsory labour ’ , as used in paragraph (2), ‘ any service of a military character ’ . There are at least two respects in which the Eritrean system of military/national service falls out with this exclusion. First of all, its legislative framework, Article 5 of the 1995 decree in particular, identifies one of the objectives of military service as ‘ to develop and enforce the economy of the national by i nvesting in development work ... . ’ The legislative framework thereby endorses the use of compulsory labour for purposes of economic development. Second, there is overwhelming evidence that in its actual practice the Eritrean state uses conscript labour for services of a non-military character. The 2015 [UN Commission of Inquiry] Report documents the use of conscript labour in construction projects and in support of private enterprise, in agriculture, in the civil service and in the judiciary ...

421. We have not found it easy to decide the issue of whether it is correct to conclude that the Eritrean system of military/national service as a whole constitutes forced labour, given that civilian national service does not ordinarily result in significant punishments and can sometimes amount to little more than attending an office in normal working hours and in the case of older women is sometimes said to be undertaken voluntarily. On balance we consider that the breach is a generic one for several reasons. First, the Eritrean government representatives before the ILO have not sought to argue that civilian national service is other than forced labour (although they dispute whether it falls within permitted exceptions). Second, ILO organs have seen it as generic. Third, even though we are unable to accept the findings of the 2016 [UN Commission of Inquiry] Report that the Eritrean system constitutes enslavement and servitude, it does particularise aspects that have a strong bearing on the issue of forced labour. Thus the 2015 [UN Commission of Inquiry] Report notes ... that;

‘ The length and conditions of work for conscripts, including wages, working hours, place of assignment, leave time and rest days do not per se constitute elements of forced labour. But the open - ended nature of national service and the often harsh working and living conditions of conscripts subjected to forced labour have a significant impact on the enjoyment of some rights including safe and healthy working conditions, the right to security, integrity of the person, and the highest attainable standard of physical and mental health ’ .

422. ... Fourth, even if not performed in oppressive conditions, civilian national service (like service in the people ’ s militia) nevertheless falls within the description of work ‘ exacted ... under the menace of any penalty ’ and also performed against the will of the person concerned, that is work for which he ‘ has not offered himself voluntarily ’ . ...

423. In relation to the exemption for ‘ any work or service which forms p art of normal civic obligations ’ (Article 4(3)(d)), we consider that the reasoning of the ILO organs applies with equal force in the context of Article 4 of the ECHR. We do not consider that the use of conscripts in civilian national service can escape the application of Article 4(3) on the basis that they form part of normal civic obligations. ... [A] s the ILO organs have consistently noted, the range and extent of work conscripts in Eritrea are required to perform in civilian national service goes well beyond anything that can be d escribed as the performance of ‘ normal civic obligations ’ , (emphasis added). The [UN Commission of Inquiry] Report of 2015 reinforces the findings of the ILO that national service is a way of controlling the population. Even though we consider discharge/release is granted more frequently than has been contended by the appellants and UNHCR, it remains that for those who have to perform such duties, the type of work a conscript is expected to do is again arbitrary and includes agricultural work, working in the mining industry and construction work. There is evidence of conscripts working for the private benefit of commanders and of the government lending conscripts to foreign companies ...

424. As regards the exemption based on provision of emergency services (Article 4(3)(c)), we consider that the ILO organs are entirely right in their repeated conclusion that the Eritrean reliance over a lengthy period on this provision goes well beyond the restricted nature of this exemption. The 2015 [UN Commission of Inquiry] Report reinforces the ILO observations, noting ... in respect of the peop le ’ s militia for example, that ‘ [T]he Commission is not aware of any such situation of emergency in the last few years that would have justified the establishment of the People ’ s Army. In any case, by definition, such situations of emergency are limited in time and compulsory labour cannot be exacted beyond the critical and genuine phase of emergency. ’

425. ... We are entirely satisfied that the open-ended duration of national service, coupled with the fact that its duration appears to be prolonged, gives rise to a real risk of a violation. There is a significant body of evidence showing that conscripts will be required to engage in work where the conditions amount to forced labour. There is strong evidence of conscripts working in the agricultural and construction industry in poor conditions. There is the Bisha mine evidence. There is strong evidence of poor conditions and mistreatment during military and some types of civic service. However, despite such evidence, we do not find that such conditions are sufficiently widespread for us to conclude that they amount to forced labour. Not all conscripts are working in conditions that would constitute forced la bour. Nevertheless, ... we consider that the lack of freedom of choice is sufficient to give rise to a breach. ...

427. For similar reasons we also consider that to the extent that the Eritrean system of military/national service breaches Article 4(2) it is also likely to give rise to a violation of Article 3.

428. We would emphasise, however, that our findings above concern active national service only. If one is a reservist subject to recall, we do not find that the risk of recall is sufficiently likely to a mount to a breach of Article 4 ...

429. We conclude that the national service regime in Eritrea does not as a whole constitute enslavement or servitude contrary to Article 4(1) of the ECHR, but that it does constitute forced labour under Article 4(3) which is not of a type permitted under Article 4(3)(a)-(d). A real risk on return of having to perform military national service duties (including civilian national service but not with the people ’ s militia) is likely to constitute a flagrant or a mere breach of Article 4(3) as well as a breach of Article 3 of the ECHR.

...

Conclusions

431. ... Although reconfirming parts of the country guidance given in MA and MO , this case replaces that with the following:

2 . The Eritrean system of military/national service remains indefinite and since 2012 has expanded to include a people ’ s militia programme, which although not part of national service, constitutes military service.

3 . The age limits for national service are likely to remain the same as stated in MO , namely 54 for men and 47 for women except that for children the limit is now likely to be 5 save for adolescents in the context of family reunification. For peoples ’ militia the age limits are likely to be 60 for women and 70 for men.

4 . The categories of lawful exit have not significantly changed since MO and are likely to be as follows:

(i) Men aged over 54

(ii) Women aged over 47

(iii) Children aged under five (with some scope for adolescents in family reunification cases )

(iv) People exempt from national service on medical grounds

(v) People travelling abroad for medical treatment

(vi) People travelling abroad for studies or for a conference

(vii) Business and sportsmen

(viii) Former freedom fighters (Tegadelti) and their family members

(ix) Authority representatives in leading positions and their family members

5. It continues to be the case (as in MO ) that most Eritreans who have left Eritrea since 1991 have done so illegally. However, since there are viable, albeit still limited, categories of lawful exit especially for those of draft age for national service, the position remains as it was in MO , namely that a person whose asylum claim has not been found credible cannot be assumed to have left illegally. The position also remains nonetheless (as in MO) that if such a person is found to have left Eritrea on or after August/September 2008, it may be that inferences can be drawn from their health, history or level of education or their skills profile as to whether legal exit on their part was feasible, provided that such inferences can be drawn in the light of adverse credibility findings. For these purposes a lengthy period performing national service is likely to enhance a person ’ s skill profile.

6. It remains the case (as in MO ) that failed asylum seekers as such are not at risk of persecution or serious harm on return.

7. Notwithstanding that the round-ups of s uspected evaders (giffas), the ‘ shoot to kill ’ policy and the targeting of relatives of evaders and deserters are now significantly less likely occurrences, it remains the case, subject to three limited exceptions set out in (iii) below, that if a person of or approaching draft age will be perceived on return as a draft evader or deserter, he or she will face a real risk of persecution, serious harm or ill-treatment contrary to Article 3 or 4 of the ECHR.

(i) A person who is likely to be perceived as a deserter/evader will not be able to avoid exposure to such real risk merely by showing they have paid (or are willing to pay) the diaspora tax and/have signed (or are willing to sign) the letter of regret.

(ii) Even if such a person may avoid punishment in the form of detention and ill ‑ treatment it is likely that he or she will be assigned to perform (further) national service, which, is likely to amount to treatment contrary to Articles 3 and 4 of the ECHR unless he or she falls within one or more of the three limited exceptions set out immediately below in (iii).

(iii) It remains the case (as in MO ) that there are persons likely not to face a real risk of persecution or serious harm notwithstanding that they left illegally and will be perceived on return as draft evaders and deserters, namely: (1) persons whom the regime ’ s military and political leadership perceives as having given them valuable service (either in Eritrea or abroad); (2) persons who are trusted family members of, or are themselves part of, the regime ’ s military or political leadership. A further possible exception, requiring a more case specific analysis is (3) persons (and their children born afterwards) who fled (what later became the territory of) Eritrea during the War of Independence.

8. Notwithstanding that many Eritreans are effectively reservists having been discharged/released from national service and unlikely to face recall, it remains unlikely that they will have received or be able to receive official confirmation of completion of national service. Thus it remains the case, as in MO, that ‘ (iv) The general position adopted in MA , that a person of or approaching draft age ... and not medically unfit who is accepted as having left Eritrea illegally is reasonably likely to be regarded with serious hostility on return, is reconfirmed, subject to limited exceptions ... ’ A person liable to perform service in the people ’ s militia and who is assessed to have left Eritrea illegally, is not likely on return to face a real risk of persecution or serious harm.

9. Accordingly, a person whose asylum claim has not been found credible, but who is able to satisfy a decision-maker (i) that he or she left illegally, and (ii) that he or she is of or approaching draft age is likely to be perceived on return as a draft evader or deserter from national service and as a result face a real risk of persecution or serious harm. While likely to be a rare case, it is possible that a person who has exited lawfully may on forcible return face having to resume or commence national service. In such a case there is a real risk of persecution or serious harm by virtue of such service constituting forced labour contrary to Article 4(2) and Article 3 of the ECHR.

10. Where it is specified above that there is a real risk of persecution in the context of performance of military/national service, it is highly likely that it will be persecution for a Convention reason based on imputed political opinion. ”

T HE LAW

I. ALLEGED VIOLATION OF ARTICLE 3 OF THE CONVENTION

54 . The applicant complained that he would face a real risk of being subjected to treatment in breach of Article 3 of the Convention if he were deported to Eritrea . Article 3 of the Convention reads as follows:

“No one shall be subjected to torture or to inhuman or degrading treatment or punishment.”

55 . The Government contested that argument.

A. Admissibility

56 . The Court notes that this complaint is not manifestly ill-founded within the meaning of Article 35 § 3 (a) of the Convention. It further notes that it is not inadmissible on any other grounds. It must therefore be declared admissible.

B. Merits

1. The parties ’ submissions

(a) The applicant

57 . T he applicant submitted that he would be arrested immediately at the airport in Asmara if he were deported to Eritrea , as the Eritrean authorities would realise that he had neither a passport nor an exit visa and was of draft age. He would then face arbitrary detention, torture , and arbitrary and inhumane punishment at the hands of either the secret service or the police owing to his desertion and illegal exit . Afterwards he would be returned to the commander of his military unit, who could also punish him arbitrarily.

58 . The applicant asserted that his account as presented to th e Swiss authorities was consistent and credible . He had provided satisfactory explanations as to the alleged discrepancies , and sufficient evidence in support of his account. There were no strong reasons to question the veracity of his account. Moreover, he was to be accorded the benefit of the doubt. Mainly repeating the submissions he had made in his appeal to the Federal Administrative Court with regard to the duration and end of his schooling, the commencement , duration and content of his military training, the date, duration and conditions of his detention , and his escape from detention, he claimed that that court had not addressed all aspects of his submissions , and had carried out a brief and superficial assessment .

59 . In particular , the Federal Administrative Court had dra w n wrongful conclusions in relation to his illegal exit. His account was concrete, substantial and reflected his personal experie nces, such as losing his way. It could not be dismissed as lacking credibility . His statements concerning the smuggler were not contradictory. They had crossed the border by crossing the river Mereb at night and, as the river was dry for most parts of the year, they had not perceive d it as anything but a rift, hence their not realising that they had crossed the border until they had been app rehended by Ethiopian soldiers. Moreover, t he documents he had submitted , notably the marriage certificate of 2010 and the baptism certificate of his son of 2012 , showed that he had been liv ing in Eritrea when he was of draft age , and that he had been registered in a refugee camp in Ethiopia in November 2013 , as confirmed by UNHCR . It was impossible for him to confirm his illegal exit by way of additional evidence , since he had left the country from an area without a border post. H e had been in good health and thus not in need of medical treatment which was only available abroad, was poor ly educated and thus unable to get a scholarship abroad , and was neithe r a businessman nor a sportsman. Consequently, it had been impossible for him to obtain an exit visa required for a lawful exit. He had therefore proved or at least established prima facie that he had exited Eritrea illegally. The Government had failed to present a plausible alternative story as to how he c ould have left Eritrea legally .

60 . The applicant asserted that the report by the State Secretariat for Migration of June 2016, to which the Government had referred in their submission s, and which had been published as the EASO report “Eritrea – National Service and Illegal Exit” in November 2016 (see paragraph s 4 6 to 4 8 above) , did not meet the standards for country information as established by the case-law of the Court in terms of independence, reliability and objectivity of sources. Relying on Saadi v. Italy [GC] ( no. 37201/06 , § 143, ECHR 2008 ) and NA. v. the United Kingdom ( no. 25904/07, §§ 119-120, 17 July 2008 ) , he argued that the sources used by that report were not sufficiently independent, reliable or objective. Most of the information came from anonymous sources. Re iterating that the Court generally exercised caution when considering reports from anonymous sources which were inconsistent with the remainder of the information before it ( Sufi and Elmi v. the United Kingdom , nos. 8319/07 and 11449/07 , § 234, 28 June 2011), he submitted that the report ran counter to the re ports of, inter alia , the Commission . T he fact-finding mission during which the information had been gathered had not visit ed detention centres or military institutions . Rather, the report was mainly based on statements by representatives of the Eritrean government and public authorities, who had an interest in portraying the situation in a positive light. Likewise, the interviews with returnees had been arranged for and conducted in the presence of government officials, who had also ensured the translation of the statements , following instructions from an official circular . The applicant made reference to one source who had made a sensitive statement contrary to that circular and had had to flee the country immediately to avoid grave repercussions. Prior to the report ’ s publication, the Eritrean authorities , which did not allow international organisations to conduct investigations in the country, had cross-checked and confirmed all the statements used.

61 . The applicant further referred to the judgment of th e Upper Tribunal of 10 October 2016 in the case of MST and Others (see paragraph 5 2 above) , and relied in particular on that judgment ’ s finding s . F irst ly , it could not be established that the payment of the 2% diaspora tax and the signing of the letter of regret would enable draft evaders and deserters to reconcile with the Eritrean authorities ( see paragraph 5 2 above, § § 333- 334 and 431 point 7 (i) of the judgment). Second ly , a person whose asylum claim had not been found credible, but who was able to satisfy a decision-maker (i) that he or she had left illegally, and (ii) that he or she was of or approaching draft age , was likely to be perceived on return as a draft evader or deserter from national service , and as a result face a real risk of persecution or serious harm ( see paragraph 5 2 above, § 431 point 9 of the judgment). The applicant argued that his illegal exit was sufficient for him to be perceived as a draft evader or deserter and to face ill-treatment as a result . He could not avoid such harm by paying the 2% diaspora tax. In the light of the Upper Tribunal ’ s findings, l ittle or no weight should be given to the report by the State Secretariat for Migration .

(b) The Government

62 . The Government submitted that the applicant mainly challenge d the Swiss authorities ’ assessment of evidence , emphasising that it was not the Court ’ s task to substitute its own assessment of the facts for that of the domestic courts, which were, as a general principle, best placed to assess the evidence before them. They reiterated that it was the applicant who had to adduce evidence capable of proving that there were substantial grounds for believing that, if the measure complained of were to be implemented, he would be exposed to a real risk of being subjected to treatment contrary to Article 3. They acknowledged that, owing to the special situation in which asylum - seekers often f ou nd themselves, it was frequently necessary to give them the benefit of the doubt in relation to assessing the credibility of their statements. Relying on A.F. v. France ( no. 80086/13 , § 46, 15 January 2015 ) , the Government submitted that , when information was presented which gave strong reasons to question the veracity of an asylum - seeker ’ s submissions, the individual had to provide a satisfactory explanation for the alleged in accuracies in those submissions .

63 . Observing that these discrepancies in the applicant ’ s account related to the duration and end of his schooling, the date of commencement and duration of his military training , the duration of his detention and the circumstances of his escape , as well as his illegal exit from Eritrea , the Government submitted that these aspects had been the subject of detailed analyses by both the State Secretariat for Migration and the Federal Administrative Court . In substance, the Government mainly repeated the arguments of the domestic authorities. In relation to the alleged illegal exit , they submitted in particular that the applicant had not explain ed the discrepancies concerning the time between his escape from detention and his departure from Eritrea , and had made contradictory statements concerning the person who had facilitated his attempt to flee . They submitted that it was hard to imagine that the applicant and the smuggler had not realised that they had crossed the border and gone in to Ethiopia at the river Mereb before being apprehended by Ethiopian soldiers , even though that river was dry for large parts of the year. The documents submitted by the applicant, notably the confirmation of his registration in the Hits as refugee camp , did not pro ve that he had left Eritrea illegally.

64 . The Government concluded by stating that the applicant had failed to clarify the significant discrepancies in his account , and that these could not be explained by the mere passage of time between his interviews with the asylum authorities. In view of the absence of substantive statements and evidence, the Federal Administrative Court could not be criticised for dismissing the applicant ’ s account as not credible. Despite being granted an additional opportunity to clarify existing doubts concerning his allegedly illegal exit in the form of a third interview , the applicant had not presented a plausible account in relation to his alleged desertion or his alleged illegal exit, distinguishing his case from that of Said v. the Netherlands ( no. 2345/02, ECHR 2005 ‑ VI ) .

65 . Referring to the information on Eritrea, notably the reports by the State Secretariat for Migration of June 2016, which had been published as an EASO report in November 2016 (see paragraphs 4 6 to 4 8 above), the Government argued that the human rights situation was not such that any removal to Eritrea would be in breach of Article 3 of the Convention. In relation to illegal exit in particular, they submitted that, in some cases, pe ople who had left Eritrea illegally were detained without being charged and could spend up to five years in detention. In other cases, however, the pe ople concerned were reassigned to national service or not sanctioned at all. In the majority of documented cases it was not possible to ascertain whether a sanction had been imposed as result of the illegal exit or for other reasons , such as desertion or disciplinary issues. They argued that an illegal exit from Eritrea was not in itself sufficient to put a person at risk of treatment in breach of Article 3 of the Convention. Notably, t his concerned pe ople who had left Eritrea prior to reaching draft age and pe ople who had left Eritrea after completing national service.

66 . In response to the applicant ’ s criticism that the report by the State Secretariat for Migration did not meet the relevant standards for country information reports, the Government pointed out that the report contained several references to sources other than the Eritrean authorities, notably independent organisations and representatives of international organisations, whose statements partly contradicted those of the authorities. All sources use d , including those which were anonymised, were evaluated in a thorough and transparent manner. The type of source was systematically indicated throughout the report , and t he credibility of the different sources was taken into consideration in the conclu sions of each chapter. Adding that t he weaknesses of the report were made transparent in the report itself , and were also reflected in the conclusions of each chapte r, the Government concluded by saying that the report was both comprehensive and balanced.

67 . In so far as the applicant alleged that part s of the report were superseded by the judgment of the Upper Tribunal of 10 October 2016, the Government contested that the respective change of practice pertained to the United Kingdom only, emphasising that neither the Norwegian nor the Swedish asylum and migration departments had expressed reservations in respect of the conclusions drawn during the peer review of the report in November 2016 in connection with its p ublication as an EASO report that month. In any event, as the applicant ’ s account as a whole had been dismissed as not credible , his submission s concerning the question of whether or not he could avoid excessive sanctio ns by paying the 2% diaspora tax were not pertinent.

2. The Court ’ s assessment

(a) General principles

68 . The relevant general principles concer ning the application of Article 3 have recently been summarised by the Court in J.K. and Others v. Sweden [GC] ( no. 59166/12 , §§ 77-105, ECHR 2016 ) .

(b) Application of these principles to the present case

69 . In accordance with the Court ’ s established case-law, the existence of a risk of ill-treatment must be assessed primarily with reference to those facts which were known or ought to have been known to the Contracting State at the time of expulsion (see F.G. v. Sweden [GC], no. 43611/11, § 115, ECHR 2016). However, if the applicant has not yet been deported when the Court examines the case, the relevant time will be that of the proceedings before the Court (see Saadi , cited above , § 133). Since the applicant in the present case has not yet been deported, the question of whether he would face a real risk of persecution upon his return to Eritrea must be examined in the light of the present-day situation.

70 . The Court notes that it is evident from the current information on Eritrea that the human rights situation in the country is of grave concern , and that pe ople of various profiles are at risk of serious human rights violations . This is also evidenced in 92% of the applications in 2016 for international protection by Eritrean nationals in Member States of the European Union plus Switzerland and Norway resulting in either refugee status or another form of protection (see paragraphs 3 6 to 5 2 above) . R eiterating that a general situation of violence would , however, only be of sufficient intensity to create a real risk of treatment contrary to Article 3 of the Convention “in the most extreme cases” where there was a real risk of ill-treatment simply by virtue of an individual being exposed to such violence on return (see, for instance, Sufi and Elmi , cited above, §§ 216 and 218 , and J.K. and Others v. Sweden , cited above, § 86, with further references), the Court notes that none of the reports conclude that the situation in Eritrea , as it stands, is such that any Eritrean national, if returned to his or her country, would r u n such a risk , nor do the reports contain any information capable of leading to such a conclusion. The Court therefore finds that the general human rights situation in Eritrea does not prevent the applicant ’ s removal per se .

71 . Hence, the Court must assess whether the applicant ’ s personal circumstances are such that he would face a real risk of treatment contrary to Article 3 of the Convention if expelled to Eritrea .

72 . The applicant claimed to be at risk of ill-treatment owing to his desertion from military service an d his illegal exit from Eritrea, things which in themselves w ere sufficient to lead to the perception that he was either a draft evader or deserter, considering that he was of draft age . The Court notes that the harsh punishment of deserters and pe ople of draft age who le ave Eritrea illegally continues to be widely reported (see paragraphs 4 0 , 4 8 and 5 2 above , §§ 281, 283, 354-355, 366-368, and 431 points 7 and 9 of the judgment ). It also notes that there are somewhat different assessments as to whether such harm could be avoided by signing a letter of regret and paying a 2% diaspora tax (compare paragraphs 4 8 and 5 2 above , §§ 333-334 and 431 point 7 (i) of the judgment ). Forcible return to Eritrea is likely to put the person con cerned at increased risk of ill ‑ treatment (see paragraphs 4 3 , 4 8 and 5 2 above , §§ 366-367 of the judgment ).

74 . The Court observes that the applicant did not submit direct documentary evidence relating to a real risk of ill-treatment which he would face in Eritrea. While this cannot be decisive per se (ibid., § 92) , it distinguishes the present case from that of M.A. v. Switzerland (no. 52589/13 , 18 November 2014) . In that case, the failure of the domestic authorities to conduct a meaningful assessment of documentary evidence relating to the alleged risk of ill-treatment in the country of origin played a crucial role in find ing that the applicant in that case had adduced evidence capable of proving that there we re substantial grounds for believing that, if expelled, he would be exposed to a real risk o f treatment contrary to Article 3 of the Convention .

75 . Reiterating that the rules concerning the burden of proof should not render the applicant ’ s rights under Article 3 of the Convention ineffective, and that it is frequently necessary to give asylum-seekers the benefit of the doubt when assessing the credibility of their statements ( J.K. and Others v. Sweden , cited above, §§ 93, 97), the Court notes that the Federal Administrative Court found that there were several discrepancies in the applicant ’ s account . The account also lacked substance and detail, notably with regard to the end of his schooling , the date of commencement , duration and content of his military training , as well as the duration and dates of his detention (see paragraph 2 8 above). The discrepancies and credibility concerns thus relate d to core aspects of the applicant ’ s claim and his account as a whole (compare and contrast N. v. Finland , no. 38885/02, §§ 154-155, 26 July 2005) .

76 . The applicant submitted that it had been impossible for him to obtain an exit visa required for lawful exit, given his age, health, level of education, and lack of involvement in business or sports (see paragraph 59 above). He relied on the country information which stated that the illegal exit of a person of draft age was sufficient for that person to be perceived as a draft evader or deserter , and consequently face ill-treatment upon a forced return to Eritrea (see paragraph 5 2 above , §§ 344-347, 354-356, 366-368, 370 and 431 points 7 (iii) and 9 of the judgment ) . In addition, he referr ed to the requirement to obtain an exit visa in order to leave Eritrea legally , and the categories of people eligible for exit visas (see paragraph s 4 1 and 5 2 above , §§ 308 , 326, 328 and 431 point 4 of the judgment ) . He also relied on his student identity card, his marriage certificate and his son ’ s baptism certificate to support his claim that he had been liv ing in Eritrea when of draft age , and added that it was impossible for him to confirm his illegal exit by way of additional evidence , as he had left the country on foot from an area without a border post. He also relied on his registration as a prima facie refugee in the Hitsas refugee camp in November 2013 to support his claim .

77 . The Court acknowledges that, in circumstances such as those claimed by the applicant, it is impossible to confirm an illegal exit from Eritrea by way of documentary evidence. It is for precisely that reason that decisive weight is attached to the plausibility of the applicant ’ s testimony . The Court notes that his account appears plausible in the light of the country information on Eritrea , and that some specific elements of his account were corroborated by the country information, notably that the Eritrean authorities initially refrained from drafting him because of his role as a church deacon , as they deferred the draft of clerics until a change in practice in 2010 led to a stricter approach (see paragraph 49 above) , and that students with poor grades were typically assigned to Wi ’ a for military training (see paragraph 5 1 above).

78 . However, the Court also notes that , in their submissions, the State Secretariat for Migration, the Federal Administrative Court and the respondent Government point ed towards a number of discrepancies and a lack of substance and detail in various parts of the applicant ’ s account, including in relation to his departure from Eritrea and other key elements of his claim. It observes that the State Secretariat for Migration heard the applicant in person three times , explicitly informed him about credibility concerns at the beginning of the third hearing , and gave thorough reason s for its assessment as to why it did not consider his account credible , in relation to his alleged illegal exit or at all (see paragraphs 19 to 2 1 above). The Court also observes that the applicant undertook to explain the alleged discrepancies in his submission s to the Federal Administrative Court (see paragraphs 2 2 to 2 7 above) . That court, in turn, gave thorough reason s as to why it did not consider his account in relation to his alleged illegal exit credible, also by drawing inferences from the concerns about the overall credibility of his account (see paragraphs 2 8 to 3 2 above) .

79 . In so far as the applicant asserted that he had credibly demonstrated that he had left Eritrea illegally , and that the Government had failed to present a plausible alternative story as to how he could have left Eritrea legally (see paragraph 59 above), the Court reiterates that it wa s for the applicant to substantiate his claim , at least as far as his individual circumstances were concerned. In that regard, the Court notes that the applicant s in the case of J.K. and Others v. Sweden (cited above) had credibly confirmed that t he y w ere victim s of past ill-treatment. The Court considered that past ill-treatment served as an indicator of future ill-treatment , and that the information available on the country concerned confirmed such a risk, concluding that it was for the respondent Government to dispel any doubts about that risk (ibid., § 102 ) . In the context of Eritrea, a similar distribution of the burden of proof may apply where it is likely – if need be by drawing inferences from the overall credibility of the person ’ s account – that a person left the country illegally despite being of or approaching draft age (see also paragraph 5 2 above, § 431 point 9 of the judgment of the Upper Tribunal of the United Kingdom in the case of MST and Others ), leaving it for the authorities to dispel any doubts about risks upon return despite those factors. However, t he shared burden of proof cannot be construed in a way which would require the authorities to pro ve that the applicant in question left Eritrea legally in each and every case, notably where the applicant ’ s overall account was not deemed to be credible . The Court shares the view s of the Upper Tribunal that a person whose asylum claim has not been found credible cannot be assumed to have left Eritrea illegally (ibid., § 431 point 5) , and that being a failed asylum-seeker is not in itself sufficient for a person to face a real risk of treatment contrary to Article 3 of the Convention upon his or her removal to Eritrea ( ibid. , §§ 335-337 and 431 point 6 ).

80 . Having regard to the above, and reiterating that the Convention system is founded on the principle of subsidiarity , and that it is not the Court ’ s task to substitute its own assessment of the facts for that of the domestic courts, which are, as a general principle, best placed to assess the evidence before them, the Court is satisfied that the assessment made by the domestic authorities was adequate, sufficiently reasoned , and supported by material originating from reliable and objective sources (see F.G. v. Sweden , cited above, § 117). It endorses the assessment by the Swiss authorities that the applicant failed to substantiate that he would face a real risk of being subjected to treatment contrary to Article 3 of the Convention if forced to return to Eritrea.

81 . Consequently, his expulsion to Eritrea would not involve a violation of Article 3 of the Convention.

II . ALLEGED VIOLATION OF ARTICLE 4 OF THE CONVENTION

82 . The applicant alleged that he would risk being subjected to treatment in breach of Article 4 of the Convention if he were deported to Eritrea . He claimed that he would be sent back to his military unit and forced to carry out indefinite military service, which, in its current state , would violate his right not to be held as a slave or in servitude and not to be required to perform forced labour.

A. The parties ’ submissions

1. The Government

83 . The Government submitted that the applicant had not exhaust ed domestic remedies in relation to this complaint , and argued for it to be dismissed as inadmissible in accordance with Article 35 § 1 of the Convention. They submitted that, before the Federal Administrative Court, the applicant had alleged a fear of ill-treatment upon his removal due to his desertion and illegal exit from Eritrea. Before the domestic authorities, h e had neither explicitly nor in substance alleged a risk of slavery, servitude and/or forced labour in the Eritrean military if he were removed to Eritrea.

84 . Referring to the case-law of the Federal Administrative Court (see paragraph 3 5 above), t he Government asserted that the applicant ’ s submission s before the Court relating to his fear of slavery, servitude and/or forced labour in the Eritrea n military constituted an examp le of a new asylum application, as the relevant facts – the details of the conditions of military service in Eritrea and its legal classification – had not been known at the time of the last domestic decision , and the new submission s would seek to establish the applicant ’ s refugee status and not only relate to impediments to the enforcement of his removal . Therefore, this did not constitute an extraordinary remedy. Consequently, t he applicant could not be exempted from the requirement of pursuing it before the domestic authorities . T he Government added that they would not speculate as to the outcome of such an application.

2. The applicant

85 . The applicant contested the Government ’ s arguments. He pointed out that he had stated in his appeal to the Federal Administrative Court that he had also fled Eritrea out of fear of having to return to the military service from which he had escaped. Acknowledging that he had not argued before that court that the military service constituted slavery, servitude and/or forced labour, he argued that he could only reasonably have be en expected to make this claim following the publication of the detailed findings of the Commission on 8 June 2016 , which had provided the necessary details about the different human rights violations inherent to military service in Eritrea. At the same time , he submitted that th is new fact was me rely a new legal classification of military service in Eritrea , and that h e had asserted the fear of being forced to return to military service in substance before the domestic authorities , and the reasons for his departure had been examined in the set of proceedings leading to the present application. A new asylum application therefore constituted an extraordinary remedy which he did not need to make use of to comply with the requirements of Article 35 § 1 of the Convention.

86 . In that regard, he also argued that i t was clear from the outset that such a new asylum application would have no prospects of success. The Swiss authorities had dismissed his account in its entirety owing to a lack of credibility. He had no new facts or evidence to present to establish the credibility of his account. Hence, there was no prospect that the domestic authorities would arrive at a different conclusion in relation to his claim that he feared being forced to perform military service in breach of Article 4 of the Convention.

B . The Court ’ s assessment

87 . The Court reiterates that , under Article 35 § 1 of the Convention , it may only deal with a matter after all domest ic remedies have been exhausted, the purpose being to afford the Contracting States the opportunity of preventing or putting right the violations alleged against them before those allegations are submitted to it (see Gäfgen v. Germany [GC], no. 22978/05, § 142, ECHR 2010). While Article 35 § 1 of the Convention must be applied with some degree of flexibility and without excessive formalism, it normally requires that the complaints intended to be brought subsequently before the Court should have been made to the competent domestic courts, at least in substance (ibid. ; see also Association Les témoins de Jéhovah v. France (dec. ) , no. 8916/05, 21 September 2010).

88 . Applicants are only obliged to exhaust domestic remedies offering reasonable prospects of success ( Scoppola v. Italy (no. 2) [GC], no. 10249/03, § 71, 17 September 2009). A mere doubt on the part of the applicant as to the prospect s of success of a particular remedy will not absolve him or her from the obligation to try it ( Epözdemir v. Turkey (dec.), no. 57039/00, 31 January 2002). Extraordinary remedies normally need not be used ( Prystav s ka v. Ukraine (dec.), no. 21287/02, 17 December 2002).

89 . The Court notes that the applicant ’ s claim as presented to the State Secretariat for Migration and the Federal Administrative Court focused on the risk that he would face ill-treatment if he were deported to Eritrea , for reasons of his alleged desertion from the military a nd his illegal exit . He did not argue before the domestic authorities that the military service constituted slavery, servitude and/or forced labour , as he has acknowledged himself (see paragraph 8 5 above) . In so far as he asserted before the Court that he had claimed to fear being forced to return to military service following his deportation, it has to be noted that the applicant stated in his appeal to the Federal Administrative Court that, following his escape from prison, he had feared that he would be detained once again or forced to perform military service, and had therefore decid ed in early October 2013 to leave the country illegally (see paragraph 2 2 above). T he Court considers that , in the set of proceedings leading to the present application, the description of this fear was primarily relevant to the description of the circumstances of the applicant ’ s departure from Eritrea , and not as relevant to the dangers he would be exposed to if he were forcibly returned (compare and contrast Kalantari v. Germany (dec.), no. 51342/99, 28 September 2000) .

90 . The Court observes that t he Federal Administrative Court rendered its decision on the applicant ’ s appeal on 9 May 2016. It considers that the applicant could only reasonably have ma de his claim that military service in Eritrea constituted slavery, servitude and/or forced labour after the publication of the second report of the Commission on 9 May 2016 . M ore importantly , he could only have done so after the detailed findings of that report, published on 8 June 2016 , had provided the necessary details about the human rights violations inherent in military service in Eritrea (see paragraphs 4 4 and 4 5 above). It also notes that country information on Eritrea which has been published since consider s that , even if a person likely to be perceived as a draft evader or deserter could avoid punishment in the form of detention and ill-treatment, he or she would likely be assigned to perform (further) national service, which would likely amount to treatment contrary to Articles 3 and 4 of the Convention (see paragraph 5 2 above, § 431 point 7 (ii) of the judgment of the Upper Tribunal).

91 . Indeed, the Court notes that the Government, referring to the case-law of the Federal Administrative Court (see paragraph 3 5 above) , submitted t hat the applicant ’ s submission s relating to his fear of slavery, servitude and/or forced labour in the Eritrean military constituted an example of a new asylum application, as the relevant facts – the details of the conditions of military service in Eritrea and its legal classification – had not been known at the time of the last domestic decision , and the new submission s would seek to establish the applicant ’ s refugee status (see paragraph 8 4 above). Therefore, the Court considers that the applicant may institute a new set of proceedings for asylum or temporary admission, in which his claim regarding Article 4 of the Convention will be examined on the merits by the State Secretariat for Migration and , in the event of an appeal , by the Federal Administrative Court. A new asylum application based on this claim would thus not constitute an extraordinary remedy.

92 . The Court reiterates that a mere doubt on the part of the applicant as to the prospect s of success of a particular remedy does not absolve him from the obligation to try it . It also reiterates the Government ’ s submission that they would not speculate as to the outcome of a new asylum application based on the applicant ’ s fear of being sent back to his military unit and forced to carry out indefinite military service, which, in its current state , would violate his right not to be held as a slave or in servitude and not to be required to perform forced labour. The Court adds that the applicant has the opportunity to lodge a new application before the Court, should such a new asylum request be rejected by the domestic authorities and courts.

93 . In view of the foregoing considerations, the Court finds that this complaint must be rejected under Article 35 §§ 1 and 4 of the Convention for non-exhaustion of domestic remedies.

FOR THESE REASONS, THE COURT , UNANIMOUSLY,

1. Declares the complaint concerning Article 3 of the Convention admissible , and the remainder of the application inadmissible ;

2 . Holds that the implementation of the expulsion order against the applicant would not give rise to a violation of Article 3 of the Convention;

3 . Decides to continue to indicate to the Government under Rule 39 of the Rules of Court that it is desirable in the interests of the proper conduct of the proceedings not to expel the applicant until such time as the present judgment becom es final , or until further order.

Done in English, and notified in writing on 20 June 2017, pursuant to Rule 77 §§ 2 and 3 of the Rules of Court.

             Stephen Phillips Helena Jäderblom Registrar President